The widespread adoption of screening mammography has led to an increase in the diagnoses of ductal carcinoma in situ (DCIS) of the breast. While it is estimated that 62,000 women will be diagnosed with DCIS in 2006, only a small fraction of these women (~10-20%) will subsequently develop invasive breast cancer or die of the disease. However, most women with DCIS are treated similarly, i.e. the DCIS is surgically excised during lumpectomy and additional radiation or tamoxifen is often offered. This means that many women are receiving unnecessary adjuvant radiation, adjuvant hormonal therapy, or mastectomy to prevent invasive cancers that will not occur. Additionally, a minority of women is not receiving adequate intervention because they subsequently develop a tumor after local excision alone. Identifying molecular markers that can accurately predict subsequent DCIS and/or invasive cancer events could aid in stratifying an individual's risk for subsequent disease and response to therapy. Thus, to avoid over- and under treatment of these women, there is a critical need for studies with complete pathology review, molecular marker measurements and long term follow-up to determine prognostic factors can be accurately and reproducibly measured and are consistently associated with subsequent tumor events among women with DCIS. Exciting preliminary data have demonstrated that expression of specific markers is associated with increased probability of subsequent tumor events following lumpectomy alone. Using a large, established and well-characterized population-based cohort of women with DCIS treated by lumpectomy alone with 8.1 years of follow-up, we will examine the role of these and additional markers in signaling subsequent disease events. To accomplish this, we will: 1) Validate assays and correlation of biomarkers with outcome retrospectively (a) using a large number of specimens (b) in different labs and (c) determine if it is statistically robust. (2) Refine clinical biomarker assay reagents using multiplex methodology. (3) Validate correlation of biomarker with outcome in a prospective human study. (4) Develop a risk assessment model to assist women and physicians with clinical decision-making about treatment for DCIS. (5) Evaluate vHMEC biomarkers as potential diagnostic targets. The risk assessment tool will estimate the risk of subsequent DCIS and invasive events for an individual woman as a function of her clinical (e.g., age at diagnosis) and histopathology (e.g., nuclear grade, margin width) information, and molecular markers (e.g., Ki67, p16, COX-2). This decision-making tool could be applied to women entering a randomized controlled trial to determine if our risk assessment tool consistently predicts subsequent disease and type of disease. The evaluation of biomarkers as potential diagnostic targets may provide the basis in the future for developing a targeted imaging and preventive agents for clinical use.